<div dir="ltr"><div><div><div>The "exact equations behind the 'probability" are<br><br> IF (s.gt.zero) then<br> z = (s-1d0)/sqrt(var)<br> ELSEIF (s.lt.zero) then<br> z = (s+1d0)/sqrt(var)<br> ELSE<br> z = zero<br> END IF<br><br> prob = erf(abs(z)/sqrt(2d0))<br><br></div>I will send you the code used offline.<br><br>====<br></div>I added the following to the documentation:<br><br></div>Example 1 ...<br><br><pre> c = (/-1,2,-3,4,-5,6,-7,8,-9,10,-11,12/)
pc = <strong>trend_manken</strong>(c, False, 0) ; pc(0)=<b>0.2683</b>, pc(1)=1.0
</pre>
<p>
For comparison, on the last example R (after loading the 'Kendall' library) returns:
</p><pre> < <b>library(Kendall)</b>
> q <- c(-1,2,-3,4,-5,6,-7,8,-9,10,-11,12)
> x <- c(1,2,3,4,5,6,7,8,9,10,11,12)
> qx <-Kendall(x,y)
> qx
tau = 0.0909, 2-sided pvalue =<b>0.7317</b> (Note: 1-0.7317= <b>0.2683</b> which matches NCL)
</pre>The NCL function does *not* return the Mann-Kendall 'tau' statistic.... just the probability (1-p)<br><div><div><div><div><br></div></div></div></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Apr 10, 2015 at 6:26 AM, Hilde Fagerli <span dir="ltr"><<a href="mailto:h.fagerli@met.no" target="_blank">h.fagerli@met.no</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
Hi,<br>
<br>
I am using ncl 6.3.0 and the function trend_manken.<br>
In the documentation (<a href="https://www.ncl.ucar.edu/Document/Functions/Built-in/trend_manken.shtml" target="_blank">https://www.ncl.ucar.edu/Document/Functions/Built-in/trend_manken.shtml</a>) it says that the output is<br>
'probability' and the 'Theil Sen trend estimate'.<br>
<br>
What is this 'probability' and how is it calculated? Can you conclude from it whether a trend exists or not? Does it take into account the MK test statistics (I have more than 10 calues)?<br>
<br>
The web page refers to <a href="http://vsp.pnnl.gov/help/Vsample/Design_Trend_Mann_Kendall.htm" target="_blank">http://vsp.pnnl.gov/help/Vsample/Design_Trend_Mann_Kendall.htm</a>, but even after reading this it is not really clear to me what this probability that is calculated refers to.<br>
<br>
I would be very grateful if anybody can help - for instance refer to the exact equations behind the 'probability'.<br>
<br>
<br>
Best regards<br>
<span class="HOEnZb"><font color="#888888">Hilde Fagerli<br>
_______________________________________________<br>
ncl-talk mailing list<br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a><br>
</font></span></blockquote></div><br></div>